Proceedings chapter

A preliminary study on the prediction of human protein functions

PublisherHeidelberg : Springer
  • Lecture Notes in Computer Science; 6686
Publication date2011

In the human proteome, about 5'000 proteins lack experimentally validated functional information. In this work we propose to tackle the problem of human protein function prediction by three distinct supervised learning schemes: one-versus-all classification; tournament learning; multi-label learning. Target values of supervised learning models are represented by the nodes of a subset of the Gene Ontology, which is widely used as a benchmark for functional prediction. With an independent dataset including very difficult cases the recall measure reached a reasonable performance for the first 50 ranked predictions, on average; however, average precision was quite low.

Research group
Citation (ISO format)
BOLOGNA, Guido et al. A preliminary study on the prediction of human protein functions. In: Foundations on Natural and Artificial Computation: 4th International Work-conference on the Interplay Between Natural and Artificial Computation, IWINAC 2011. Heidelberg : Springer, 2011. p. 334–343. (Lecture Notes in Computer Science)
Main files (1)
Proceedings chapter (Published version)
  • PID : unige:108944

Technical informations

Creation09/15/2018 1:38:00 PM
First validation09/15/2018 1:38:00 PM
Update time03/15/2023 8:44:54 AM
Status update03/15/2023 8:44:53 AM
Last indexation05/02/2024 9:22:06 PM
All rights reserved by Archive ouverte UNIGE and the University of GenevaunigeBlack